Upload tokenizer
Browse files- vocab.json +43 -43
vocab.json
CHANGED
|
@@ -1,51 +1,51 @@
|
|
| 1 |
{
|
| 2 |
-
"'":
|
| 3 |
-
"L":
|
| 4 |
-
"N":
|
| 5 |
-
"[":
|
| 6 |
"[FY-NL]": 47,
|
| 7 |
"[NL]": 48,
|
| 8 |
"[PAD]": 46,
|
| 9 |
"[UNK]": 45,
|
| 10 |
-
"]":
|
| 11 |
-
"a":
|
| 12 |
-
"b":
|
| 13 |
-
"c":
|
| 14 |
-
"d":
|
| 15 |
-
"e":
|
| 16 |
-
"f":
|
| 17 |
-
"g":
|
| 18 |
"h": 0,
|
| 19 |
-
"i":
|
| 20 |
-
"j":
|
| 21 |
-
"k":
|
| 22 |
-
"l":
|
| 23 |
-
"m":
|
| 24 |
-
"n":
|
| 25 |
-
"o":
|
| 26 |
-
"p":
|
| 27 |
-
"q":
|
| 28 |
-
"r":
|
| 29 |
-
"s":
|
| 30 |
-
"t":
|
| 31 |
-
"u":
|
| 32 |
-
"v":
|
| 33 |
-
"w":
|
| 34 |
-
"x":
|
| 35 |
-
"y":
|
| 36 |
-
"z":
|
| 37 |
-
"|":
|
| 38 |
-
"á":
|
| 39 |
-
"è":
|
| 40 |
-
"é":
|
| 41 |
-
"ê":
|
| 42 |
-
"ë":
|
| 43 |
-
"ï":
|
| 44 |
"ó": 18,
|
| 45 |
-
"ö":
|
| 46 |
-
"ü":
|
| 47 |
-
"–":
|
| 48 |
-
"’":
|
| 49 |
-
"“":
|
| 50 |
-
"”":
|
| 51 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"'": 39,
|
| 3 |
+
"L": 15,
|
| 4 |
+
"N": 38,
|
| 5 |
+
"[": 25,
|
| 6 |
"[FY-NL]": 47,
|
| 7 |
"[NL]": 48,
|
| 8 |
"[PAD]": 46,
|
| 9 |
"[UNK]": 45,
|
| 10 |
+
"]": 19,
|
| 11 |
+
"a": 8,
|
| 12 |
+
"b": 2,
|
| 13 |
+
"c": 3,
|
| 14 |
+
"d": 12,
|
| 15 |
+
"e": 35,
|
| 16 |
+
"f": 11,
|
| 17 |
+
"g": 24,
|
| 18 |
"h": 0,
|
| 19 |
+
"i": 42,
|
| 20 |
+
"j": 31,
|
| 21 |
+
"k": 9,
|
| 22 |
+
"l": 29,
|
| 23 |
+
"m": 30,
|
| 24 |
+
"n": 34,
|
| 25 |
+
"o": 27,
|
| 26 |
+
"p": 7,
|
| 27 |
+
"q": 21,
|
| 28 |
+
"r": 5,
|
| 29 |
+
"s": 10,
|
| 30 |
+
"t": 22,
|
| 31 |
+
"u": 23,
|
| 32 |
+
"v": 6,
|
| 33 |
+
"w": 17,
|
| 34 |
+
"x": 32,
|
| 35 |
+
"y": 28,
|
| 36 |
+
"z": 13,
|
| 37 |
+
"|": 37,
|
| 38 |
+
"á": 16,
|
| 39 |
+
"è": 43,
|
| 40 |
+
"é": 20,
|
| 41 |
+
"ê": 41,
|
| 42 |
+
"ë": 33,
|
| 43 |
+
"ï": 14,
|
| 44 |
"ó": 18,
|
| 45 |
+
"ö": 26,
|
| 46 |
+
"ü": 1,
|
| 47 |
+
"–": 40,
|
| 48 |
+
"’": 44,
|
| 49 |
+
"“": 36,
|
| 50 |
+
"”": 4
|
| 51 |
}
|